
1 / 1
【Overview】 AGENTS.md is a simple, open Markdown format for giving AI coding agents project context and instructions. Think of it as a README written for agents rather than humans: a dedicated, predictable place in a repository where you document build steps, tests, and conventions so that coding agents can work on your codebase effectively. It is used by over 60k open-source projects. The format solves the problem of proprietary, tool-specific configuration files. Rather than introducing another proprietary file, AGENTS.md uses a name and format that can work for anyone, making it easy to share instructions across different AI coding agents. It is designed for developers building or using AI coding agents, maintainers of open-source projects, and teams with large monorepos. AGENTS.md is free to use and straightforward to adopt. You add a Markdown file to your repository and fill it with the context agents need: how to build the project, which tests to run, code style guidelines, and commit message conventions. Agents read the nearest AGENTS.md file in the directory tree, so you can tailor instructions per package in a monorepo. 【Key features】 • Agent instruction file format: Provides a dedicated, predictable place in a repository to give coding agents context and instructions such as build steps, tests, and conventions. • Nested per-package instructions: Lets large monorepos place another AGENTS.md inside each package so agents read the nearest file in the directory tree. • Automated test execution guidance: Agents attempt to run relevant programmatic checks listed in AGENTS.md and fix failures before finishing a task. • Migration via symlink: Supports renaming existing docs like AGENT.md to AGENTS.md with symbolic links for backward compatibility. 【Use cases】 1. Giving AI coding agents project context and instructions. 2. Documenting build and test commands for agents. 3. Specifying code style guidelines for agents. 4. Providing commit message and pull request guidelines. 5. Shipping tailored instructions per subproject in a monorepo. 【Pros and cons】 Pros: - Opens agent instructions to any tool, avoiding lock-in to a single vendor's proprietary format. - Simple Markdown means low learning curve and easy adoption across teams. - Nested files give monorepo maintainers fine-grained control without duplicating instructions. Cons: - The closest AGENTS.md to the edited file wins, and explicit user chat prompts override everything, so instructions may be superseded by user input. - Automated test execution guidance only works if you list the testing commands in the file. 【FAQ】 Q: How does AGENTS.md work with existing files like AGENT.md? A: You can rename existing docs like AGENT.md to AGENTS.md and use symbolic links for backward compatibility. Q: Can I have different instructions for different packages in a monorepo? A: Yes, you can place another AGENTS.md inside each package, and agents will read the nearest file in the directory tree. Q: What happens if a user gives a chat prompt that contradicts AGENTS.md? A: Explicit user chat prompts override everything, so the agent will follow the user's instructions over the file.
free
WPS AI is an AI assistant built into WPS Office that helps users write documents, generate presentation outlines and…
Baidu Comate is an AI coding assistant built by Baidu on its ERNIE model. It goes beyond simple autocomplete: it…
UniConverter is an AI media conversion and enhancement tool for video, audio, and images. It combines everyday…
Added
10/22/2026
Views
659
Clicks
2
6 similar tools to consider instead of AGENTS.md